Over five million Americans have Alzheimer?s disease (AD) today. A critical goal of biomedical research is establishing indicators of AD during the preclinical stage (i.e. biomarkers) allowing for early diagnosis and intervention. Currently, the relationship between A? pathology and neuroimaging/cognitive measures during the preclinical stage is not well understood. Our project combines novel high-resolution magnetic resonance imaging (MRI) tools, novel cognitive testing particularly sensitive to hippocampal memory, and assessments of beta-amyloid (A?) pathology in cerebrospinal fluid (CSF) or florbetapir (18F) PET scans to gain a better understanding of the neural basis of preclinical AD and to identify novel mechanistic biomarkers for preclinical AD using non-invasive techniques. The goals of the project are to (1) improve our ability to detect subtle cognitive decline, (2) enhance the sensitivity of standard neuroimaging biomarkers for preclinical AD as well as develop and validate novel biomarkers using ultrahigh-resolution neuroimaging techniques, (3) test the validity of biomarker candidates in two special cohorts as models of early susceptibility (Down syndrome ? DS) and late resistance (nondemented 90+), and (4) deliver a proof of concept high-resolution multimodal neuroimaging platform for the ADRC to build the infrastructure necessary for establishing a high-resolution neuroimaging core. We will recruit and test a total of 90 participants: (a) Longitudinal cohort: 30 healthy nondemented participants (15 A?+ and 15 A?-) and 15 A?+ amnestic MCI participants between the ages of 65 and 85; (b) Down syndrome cohort: 15 A?+ nondemented individuals between the ages of 45 and 65; (c) 90+ cohort: 30 nondemented 90+ participants (15 A?+ and 15 A?-). A? status will be determined for longitudinal cohort via CSF (through ADRC Path Core) and for DS/90+ via florbetapir PET (through synergistic NIH R01?s by Dr. Kawas and Dr. Lott). In Aim 1, we will use a set of newly developed cognitive tests of pattern separation. These tests vary mnemonic interference in the object, spatial and temporal domains. In Aim 2, we will collect high resolution structural MRI (0.55 mm isotropic), resting state fMRI (1.5 mm isotropic) and DTI (0.66 mm inplane) to test hypotheses about changes in neural features related to pathological status (e.g. entorhinal cortical thickness, perforant path integrity, resting state connectivity between the entorhinal cortex and the hippocampus). In Aim 3, we will apply the techniques in Aims 1 and 2 to Down syndrome and 90+ participants to explore resistance and vulnerability to pathology in this brain network. In Aim 4, we will examine relationships between our biomarkers and pre-existing measures in the ADRC database such as genetics, blood/serum markers, and other CSF pathologies (e.g. phospho-tau). In Aim 5, we will build the infrastructure to archive and curate all new imaging data for dissemination to ADRC investigators to provide a proof of concept for a future neuroimaging core at the ADRC. Together, the Aims of this project will allow us to better understand the condition of preclinical AD and develop biomarkers that can be used in future prevention trials.